One like me! Toying with the Doll Industry


Dr. Sian Jones is a Teaching Fellow at Goldsmiths, University of London. Her research focuses on discrimination and prejudice among children and adults based on membership of a given group – and how friendships may be encouraged between children from different groups. Here, she looks at the Psychology behind the importance of representing disability in the toy industry. 


A lot of attention has focused on the toy industry of late, alongside changes in what is available and who it is targeted at. This ranges from the “let toys be toys” campaign pressuring for non-gendered marketing of products, to a plethora of companies like this one  marketing toys specifically designed to eradicate ethnic bias in dolls. This is coupled with changes to Barbie dolls both to make their shape more realistic, and to represent the careers that women may pursue.

Another avenue of change has been led by the #toylikeme campaign, with a recent petition garnering over 800 000 signatures, pressuring the industry to represent disability in their product line. Watching the video below, one could accuse anyone not able to see the joy of representation of being utterly cold-hearted.


Anecdotally, I know that my mother, and many others, banned Barbies from my toy collection . This only served to make them more attractive to me, in spite of not really being a “doll-playing” child  ( I persuaded a friend to gift me one). And I can’t tell you how I would have reacted to a toy like the one above, because I never had one. But – as a scientist – I know that the plural of anecdote (be it from my own experience, or the video above) is not data – not evidence.  So are the  changes in the toy  industry worhwhile psychologically? Are they grounded in research evidence?  Or should we retire the representative toy box to the attic?

A review of the evidence suggests not. Firstly,  it turns out that what a doll looks like matters. Ditmar (2006) took 162 girls, from age 5 to age 8, and exposed them to images of either Barbie dolls, Emme dolls (U.S. size 16), or no dolls (baseline control). Girls exposed to Barbie reported lower body esteem and greater desire for a thinner body shape than girls in the other  conditions. Moreover, the careers that dolls are depicted as having does have a real influence on children’s conceptions of the career paths that are open to them (Coyle, 2010). Coyle took girls aged 4-5 years, and interviewed them, using Barbies dressed in gender traditional and non-traditional career outfits. She found that girls identified more often with the dolls in gender traditional career clothes. However, following exposure to the  dolls with gender non-traditional careers, girls saw themselves being able to engage in a greater proportion of gender non-typical careers as an adult than before exposure. This study has since been replicated (see Sherman & Zubriggen, 2014, for example) with consistent findings.

The Seminal Doll Studies 

Next, let’s look at the question of ethnicity. This is actually where the central debate started, with a study by Clark & Clark (1947) that has become a classic in Psychology.


In what became a series of experiments, Clark and Clark showed Black children between the ages of six and nine years two dolls, one white and one black, and then asked them questions, like “show me the doll that you like best or that you’d like to play with,”or “show me the doll that looks ‘bad’.” In these studies,  around 44% said the White doll looked like them. In one version of the study Clark gave the test to 300 children in different parts of the country. He found that Black children who went to segregated schools were more likely to pick the white doll as the nice one. These findings carried considerable weight. They became part of Brown v Board of Education case. And they began the start of a movement towards multiracial education. A study by Powell-Hopson and Hopson (1988) showed that in a post-intervention aimed at tackling negative stereotypes, children chose a black doll. And like the gender / career studies, this one has also been replicated, ad infinitum.  For example, in 2005, Davis asked 21 children and 71% told her that the white doll was the nice one.

Dolls are not us

In spite of the compelling findings above, the eagle-eyed among you may have started to critique them. Recent replications of the Clark doll test have used very small samples, and were not published academically. And, it has also been noted that the contact children had with different ethnicities was not controlled, and the study lacked a control for ethnicity, of both the dolls, and the experimenters. Moreover, children don’t actually play with the dolls, and this is a forced-choice task.

In the most critical appraisal, Bergner (2009) noted that these studies show very little beyond preference: they have been used to make claims about self-esteem, but self-esteem has not actually been measured. Further, since Black children do have commensurate self-esteem to white children these preference studies are just that – about preference – representing the political and capitalist culture of the time. Wanting that to change, requires much more than raising self-esteem among Black children.

In sum, there is little in the ethnicity  “doll studies” to suggest that there is a psychological need for investment in representative toys to boost children’s self-esteem. But that is because there is no evidence – not because the evidence suggests otherwise.

Back to the toy box

Before we grow disilusioned, and throw out the doll with the bath water, let’s return to disability. Here, the  toy box is less replete with resources. The market simply doesn’t prodce so many toys – although Playmobil and Lego are  due to bring out such examples this year. “Doll” study findings are consistent with the above (and the above caveats). For example, Saha et al. (2014)  showed children with Down Syndrome two dolls, one with a “typically developing” appearance and one with the phenotypic features of Downs Syndrome. Fifty four children participated in play sessions with both dolls and were then interviewed. They prefered to play with, and attributed more positive traits to, the typical doll than the doll with Downs Syndrome.

That’s all very well, but, as noted earlier, it takes two to tango. And going back to the market, isn’t it important that toys represent diversity to all  children? In a study by Srinivasan and  Cruzhis (2015) children between 6 and 13 years of age used ethnically diverse dolls to explore and verbalise their knowledge of ‘race’. Children were able to articulate how they related to these dolls. Dolls, it seems, may be an important educational tool, for opening up dialogue. Further studies with such dolls (e.g., Smith, 2013) indicate that such dolls support the development of young children in increasing their empathy and in opening discussion about treatment of stigmatized groups. A project evaluating their use is underway.

Where now?

It seems from this review that – as far as gender is concerned – there is  strong evidence for the psychological impact of (poor) representation on children’s Psychology. There is less evidence of impact from ethnic doll studies – but this is largely because the studies have measured only ethnic preference, and not looked at outcomes like career aspirations or race-related self-esteem. Studies that have, have noted an impact. The research agenda might turn to closing this gap: to measuring the self-image, aspirations, empathy and anxiety (as well as prejudices) of those  from all groups exposed to representative toys. As it stands, in a world of White privilege, it is arguably hardly surprising that a white doll is preferred.

When it comes to disability, unlike with ethnicity, one finds that such research has not been an option until very recently: the toys themselves are, relatively speaking, all-but absent.  We know that children most easily discuss issues in contexts that are familiar to them – such as play (Srinivasan &  Cruzhis, 2015). In this regard, there is growing evidence that representative dolls are a good thing. They open dialogue around prejudice and enable discussion and empathy. If such toys are not there, the opportunity for this discussion is lost. At a broader level, if we do want to change the status quo in our society – we know that acceptance of prejudice and inequality doesn’t magically appear at 18 years of age. When it comes to fairer representation, there might be no better place to start than the toy box.